Evaluation of Metallic Artifacts Caused by Nonpenetrating Titanium Clips in Postoperative Neuroimaging

Abstract

Nonpenetrating titanium clips create no suture holes and thereby reduce cerebrospinal fluid leakage after dural closure. However, no data exist regarding metallic artifacts caused by these clips during postoperative neuroimaging. We aimed to evaluate clip-related artifacts on postoperative magnetic resonance (MR) images of 17 patients who underwent spinal surgery. A phantom study evaluated the size of metallic artifacts, and a clinical study evaluated the quality of postoperative spinal MR images. Both 1.5-T studies used T1-weighted and T2-weighted fast spin echo sequences. The phantom study compared clip and artifact size for 10 clips. Artifacts were defined as signal voids surrounded by high signal amplitude that followed the clip shape. In the clinical study, 2 neurosurgeons assessed 22 images from 17 patients of the spinal cord, cauda equina, and paravertebral muscles adjacent to the nonpenetrating titanium clips, using 5-point scales. Mean metallic artifact sizes were 4.82 ± 0.16 mm (T1) and 4.66 ± 0.25 mm (T2; P < 0.001 vs. control). The former and latter were respectively 207% and 200% larger than the clip size. Both readers graded spinal cord and paravertebral muscles images as 3 or 4, indicating very good image quality regardless of clip-related artifacts, with excellent interobserver agreement (κ= 0.99 and 0.98, respectively). Metallic artifacts caused by nonpenetrating titanium clips were 200% larger than the actual clip but did not affect spinal cord and extradural tissue visualization. The use of these clips for closing the spinal dura mater does not alter postoperative radiologic evaluation quality.